#MW17-TN Workshop Notes

#MW17-TN Workshop Notes (April 20, 2017) – Online Collections Community of Practice Workshop

What did we do? Workshop participants sat at tables that had one of four placecards: purpose, internal challenges, external challenges, and success. These placecards focused the discussion on a particular area of online collections issues. Following an initial discussion on the “big issues” of these areas, the groups talked about some more specific issues and questions, and then shared with the room as a whole. Once everybody had the chance to share, each table was asked to decide on up to three “hot topics” that they saw as the most important things, or aspects that multiple tables spoke about.

Subtopic 1: Purpose (Strategic)

Table 1:

Expectations:

  • Internal
    • Obligation (we’re supposed to do it)
    • “It’s what people in the 21st Century do” “be relevant” “be on the interwebz” “it’s easy”
    • Present the “Collection”
    • Be the base layer of access to collections, not the only one
    • It is for external people only
    • It is for niche groups only
    • Spectrum of expectation and understanding of usefulness
    • It’s in our act
    • Stakeholders, traditional understandings
    • To grow audiences
    • Digital “fixes things”
    • Increases Sales, donors, members
    • Tombstone, image, search, advanced search
    • Represent the collection in the way we understand them, eg the difference between archives and photographs, library and art
    • Be their own pinterest
    • Convert people from collection researchers to visitors, and visitors to collection users
    • The collection is what we know, is our opinion
  • External
    • I want access to your collections, and use the info
    • Stories, as well as objects
    • Browse collections, find similar, related collections
    • Interested in what they are interested in, not the entire collection
    • They want to share
    • Range of expectations of information levels, from canonical sources, be able to reference, through to simply seeing images
    • Big images on everything
    • Half decent/complete data on everything
    • The online collections is the “truth” Mission/Scope/Core Principles
    • Open access
    • Make the collections accessible, in the best form we can, as complete as possible
    • Transparency
    • There are some items we don’t make available for various reasons
      • Unknown or dodgy provenance
      • Potentially offensive, human remains, cultural sensitivities, nudity, ecologically
    • Continue social or organisation mandate through online collections, share cultures, histories
    • Connecting people to their “objects”, then maybe enable community to add context, reuse etc.

Similarities: We have related content across museums, galleries, no institution has everything on any subject, place, person

Not expectation, but came up in conversation: Staff do use it instead of CMS

Table 2:

  • Is the question Why do we have an online collection?
  • There is an academic mission to make our collection available for researchers, collaboration – only 1.5% of our collection is on view at a time
  • Devil’s Advocate: there are online collections in Europe, it seems like the museums there don’t know why they’re doing it
    • Sometimes they end up with hundreds of records in their online collections that are useless
    • Search facilities are okay, but not great – difficult to navigate and discover
    • What’s missing? Are the highlights there?
  • In the US, the online collection is a marketing tool
    • Also a tool to increase your audience to people who aren’t in your community
  • Online collection is integrated into our library search, too
  • Online collections is different from online resources
    • Just tombstone information is not enough
    • Not connected to the resources
    • List of artworks is not great if there’s no context
    • Mimics internal database and all the relationships (artists, objects, exhibitions, events, publications, including citations, media coverage, installation views)
  • What is an online collection is the better question.
  • Convert online articles to PDFs and link to objects/events are multimedia
  • How do you contextualize the object in your collection to other online resources or objects in other collections, to enable teaching?
    • Academic researcher providing those connections
    • Trying to get those connections out of the academics’ heads
    • There are other audiences, too — not just researchers
    • People come via Google, visit what they’re interested in, and leave
  • At a basic level, they need to be responsive/mobile, accessible
  • Should online collections be archival? Are they archives?
  • Online collections are good at telling stories, e.g., restoration of objects
  • Interest by faculty in integrating objects into teaching
  • How to circle those stories back into the online collection?
  • Audience that you serve → for university museums, sticky audience with lots of repeat use
    • For more general audience institutions, less focused on research and more on general interest (perhaps not as deep in engagement)
    • K-12 audience is really big for university museums as well
  • Clear that the goal of online collections is to serve our communities
  • Not just about providing information but also telling stories and providing context
    • Ideally feed stories back into the collection to make it richer
  • Need to have entryways as online collections get larger
  • Don’t let perfect be the enemy of the good
  • For expectations, audience research is important
  • Accuracy? AIC wants to be sure that data published is accurate
    • Most people send constructive criticism – they do not feel it reflects badly on the institution
  • Putting it out in the public is good for grants ($$$)

Reported Summary:

  • 2 sets of expectations – internal/external
    • What organizations believe they are doing when they put a collection online, and what audiences think they are doing
      • Internal – we have to do it because everyone else is going it; access to others; that it is only for niche or distance users; to grow audiences; digital “fixes” things; to increase tickets, donors, members, sales; expected to put out image, info, search, advanced search; to represent the collection how we understand to represent it; create our own pinterest; to increase interest and convert users to citizen scientists
      • External – find what they are interested in; share; be able to reference; find a cool image; bigger images and complete data on everything; online collections as truth
    • Mission – open-access as part of mandate; aim for (some) transparency, with consideration for sensitivities; connecting people to content
    • Similarity – we hold similar content, not institution can claim to tell entire story about one particular area/subject
  • Goal is to serve our communities
  • Don’t let “perfect” data get in the way – it is okay to not be perfect
  • Expectations – audience research informed what we think those expectations are

Subtopic 2: Internal Challenges (Tactics)
  • Relational vs. “Flat” data model
  • Internal discussions about how records are to be represented (eg. “tea set” – one record vs. record hierarchy)
  • Consensus on cataloguing standards (ie. “What is this thing?”)
  • How much data do we publish online?
  • Which system is “system of record” (DAMS vs. Content Management Systems vs. Collections Management System)
  • Working with 3rd party developers (proprietary database system vs. open system, limited pool of web developers that work with museums, limited pool of web developers willing to work with in-house technology)
  • What resources do we have available (e.g images of collection objects) – How much content do we need to create?
  • Resources – What skills are we looking for when hiring resources for web implementation, if existing, do we backfill that position temporarily

Reported Summary:

  • CMS does not always support fast/responsive websites
  • How to we get consensus on cataloguing standards
  • How do we choose what to publish
  • Which system to we publish from
  • Challenge of working with third party supporters
  • Proprietary software issues
  • How do we make sure we have all the resources we need to put collections online
  • Categories:
    • 1. How to we facilitate internal discussions to decide on what data we are publishing? Are we even ready/how do we get ready?
    • 2. Systems – how do we decide which to use; how do they interoperate; are they sustainable/interoperable
    • 3. Resources – how do we make sure we have what is necessary
    • 4. Risk tolerance and copyright licensing

Subtopic 3: External Challenges (Effects/Outputs)

Table 1:

  • 1.Discoverability and Social Media
    • SEO
    • Social Media Campaigns
    • Do you want them to go back to website or push content to social platform
    • Do we want to push our online collection to social platforms or do we only want to use social platforms as a way to drive traffic to our website where they’ll be exposed to our website?
    • Do you segment audience by their preferred platform
    • Do we push it to the platform they like or invite them to our website
    • Is the ability to share to social media outlets a requirement?
    • Search across all platforms is important because we want people to discover not just the work of art but everything we have related to that work of art – programs publications
  • How do you decide what part of your collection you put on-line?
  • Are there standards for what you share – more institution based
  • Need to come to the realization that the work will never be done , always need to fine tuned , not to mention new things that come into the collection.
  • IIIF – interoperability standard  – format for uploading images so it can be used across institutions
  • Is it OK to press forward if everything isn’t perfect – like if data not 100% accurate YES! – but will our audience accept that ?
  • How much do we share with others – do we put all our collections  data out on github – do we build out an API?
  • Many Audiences
  • Standards for sharing
  • Open Data & Licensing

Table 2:

  • What would be success, externally?  Need to understand the scope to know what the problems are to get there.
  • Audience: [Very high level]
    • Visit duration / engagement metrics. Was the content useful. Was it easy to say it’s not useful.
    • Measuring/tracking use, and type of use.
    • Audience is everyone!
  • Discoverability:
    • Data discoverable … but maybe only for certain audiences/uses?
    • Visualization of the information important to not require knowledge of the collection
    • Different success criteria for different audiences.
    • Discovery is manageable? (updates, publishing, API not union catalog)
    • Cross-institution integration, such as recommender systems that can bring in others’ data or provide a link to other collections
  • Inclusion:
    • How to make the information inclusive externally for various communities. NB audience is everyone.
    • Accessibility
    • Responsive
    • International
  • Usage / Usability:
    • Can reuse content at all? (api)
    • Knows how can use (license)
    • Tracking use difficult
  • Sustainability:
    • Extensible?
    • Separation of data management, access, presentation and discovery  
    • Front end as disposable, data as core value
    • Managing change at the API/model/data level

Reported Summary:

  • What would success look like from an external perspective?
    • Discoverability for different audiences
  • Cross-institution integration
  • Accessability
  • Usage and usability – can the content be reused at all? Can you access the data ie via API?
  • It is okay to press forward if things aren’t perfect – but is that actually acceptable to our audiences?
  • Do we want to push our collection to social platforms to attract visitors to website, or to see SM as new learning platform?
  • The github, API question

Subtopic 4: What does Success Look Like (Measurements/Evaluation)

Table 1:

A meta theme: what is meaningful impact?

  • Sensible metrics
    • UI engagement: appropriate event analytics
      • How are people engaging on the object page? Are they clicking on content accordions, for example?
      • Heat maps (hards to understand cross object)
    • Big picture metrics
      • Visits / image downloads / search criteria
    • Cohort analytics
      • Provide an attractive option for account creation / self identification?
      • Support for external users with different access levels?
    • Integrated social media analytics
      • Usage of this asset on tw, fb, etc.
    • Internal metrics // internal content updates
    • Data normalization // success around record completeness?
  • Increased visibility
    • Story layer vs. data layer: data is easy to put out;
    • Clearinghouse resources of catalogues?
  • Increased integration
    • Open data, open API? API keys?
    • Who is referring in?  Eg., Pinterest.
    • Distribution? DPLA?
    • Open data: CSV data published via github? (techie audience)
  • Increase sharing of tools
  • A community of institutions using a given tool?

Table 2:

  • Sensible metrics:
    • Agreed upon
    • Know up front expectations
    • How is data comparative – how is it comparable between museums?
  • Internal buy-in:
    • Systems are usable → sustainability beyond individuals and/or positions
  • Sustainable:
    • Not privilege based
    • Integration of systems
  • Accessibility
  • Funding – availability
  • Increased visibility
    • Audience
  • Increased sharing
    • Expansion of data set use
    • Known or unknown

Reported Summary:

  • Metrics as internal and external
    • Internal – content updates; data normalization; success around record completeness
    • External – UI engagements; event analytics; heat/usage maps; big picture metrics
    • Cohort analysis – supporting external users with different levels of access and using these to develop cohort sets for analysis
  • Distribution – online catalogues and libraries
  • Integration – CSV on Github, but not highly technical audience that can use it
    • APIs with API keys – gives us better sense, thing to concretely measure
  • Open-source tools – identify community of tool users
  • Sensible metrics – process of starting to put a collection online; metrics have to be agreed upon at the beginning even if you don’t have a baseline; want to be able to compare metrics and results across institutions
  • Increased visibility – audience numbers; increased sharing is a way to measure success although you sometimes don’t know what is being done with it/how it is being used; expansion of use
  • Integration – important for internal systems within the organization
  • Internal buy-in essential
  • Sustainability
  • Accessibility – describe your images! Anything is better than nothing; language issues
  • Funding as a measure of success (coming in under budget?)
Hot Topics
  1. Common metrics // sharing metrics across institutions // common languages of engagement?
    1. Data – How do we facilitate internal discussions to reach agreement on what data we are publishing, how we are representing records (ie. data/cataloguing standards), and whether or not we are ready?
  2. Effective institutional data sharing?
    1. Discoverability through collaboration (data sharing)
  3. Internal buy-in
    1. Connect, across collections, across org
  4. Resources – How do we ensure we have the necessary resources to achieve our online collections publication strategy (project management, IT translator, financial, content generation, vendor management, etc.)?
    1. Technical knowledge
    2. API development and use
  5. Accessible by default, but there are expectations, how do we find that line?
    1. Audience expectations don’t necessarily align with audience expectations. Do we know what audience expectation. Particularly in relation to the breadth of the collections. The data, images, structures are the core, the products that should be…?
    2. Relevant (to audiences)
    3. Open (including imperfection)
    4. Used (usefulness proved by use)
  6. Systems – How do we decide which systems are part of our online collections publication strategy, how they interoperate, and are they sustainable and supportable?
  7. Rights – How risk averse should an institution be when publishing images online (restrictive rights policy vs. fair use licence)?
  8. Survey of where we’re at, gaps, what hindrances are there to progress?
  9. Sustainability

Going Forward – Areas to watch/join

www.museweb.wiki – updates, articles, and resources on all CoPs; sign up!

mwcop.slack.com – Slack channel with discussion and access to online collaborative working spaces

#mwcop – find us on Twitter using this hashtag, join and start conversations